Methanol and Ethylene Glycol Poisoning: a Case Study and Review of Current Literature

TOXICOLOGY • TOXICOLOGIE

Methanol and ethylene glycol poisoning: a case study and review of current literature

William R. Henderson, MD; Jeffrey Brubacher, MD

ABSTRACT Poisoning is an uncommon but potentially fatal outcome of toxic alcohol ingestion. The toxic alcohols methanol, ethylene glycol and isopropyl alcohol are commonly found in household and commercial products. Because the toxic effects are caused by the metabolites of methanol and ethylene glycol rather than the agents themselves, there is often a substantial delay between ingestion and onset of clinical toxicity. Anion and osmolar gaps are often used for the diagnosis and exclusion of these sometimes subtle overdoses. The pitfalls of using these tests to rule out alcohol ingestion are reviewed. Ethanol infusion is the traditional therapy for such overdoses. In addition to the pathophysiology and clinical findings in poisoning, recent evidence for the use of fomepizole and adjuvant therapies is reviewed.

RÉSUMÉ L’empoisonnement est une conséquence rare mais potentiellement fatale de l’ingestion d’alcool toxique. Les alcools toxiques comme le méthanol, l’éthylène-glycol et l’alcool isopropylique sont très répandus dans les produits ménagers et commerciaux. Comme les effets toxiques sont causés par les métabolites du méthanol et de l’éthylène-glycol plutôt que par les agents eux-mêmes, il existe souvent un délai important entre l’ingestion et la toxicité clinique. Les écarts anioniques et osmolaires sont souvent utilisés pour le diagnostic et l’exclusion de ces intoxications parfois sub- tiles. Nous présentons les pièges associés à ces tests utilisés pour exclure l’ingestion. La perfusion d’éthanol est la thérapie couramment utilisée pour traiter ces intoxications. En plus de la phys- iopathologie de l’empoisonnement et des constatations cliniques, des données récentes suggérant le recours au fomépizole et à des thérapies d’appoint sont passées en revue.

Key words: methanol, ethylene glycol, toxicology, fomepizole, osmolar gap, alcohols, poisoning

Introduction should be considered when a patient presents with a history or clinical picture consistent with such ingestion, asso- The toxic alcohols methanol, ethylene glycol and isopropyl ciated with acidosis and serum chemistry with a widened alcohol are widely available in household and commercial anion or osmolar gap. products that are intentionally abused as ethanol substitutes. Although isopropyl alcohol is relatively benign, both Case report methanol and ethylene glycol have serious, sometimes fatal, effects. Because of the widespread availability of these Emergency medical services were called to a location agents and their potential toxicity, toxic alcohol ingestion where a 38-year-old man was performing cardiopul-

Department of Emergency Medicine, Vancouver Hospital and Health Sciences Centre, Vancouver, BC Received: Feb. 4, 2001; final submission: Oct. 20, 2001; accepted: Oct. 27, 2001 This article has been peer reviewed.

34 CJEM • JCMU January • janvier 2002; 4 (1) Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.93, on 26 Sep 2021 at 20:37:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1481803500006035 Methanol and ethylene glycol poisoning

monary resuscitation on his wife. The woman was asys- Methanol tolic and unresponsive to all resuscitative efforts. The man became increasingly agitated and incoherent. The Clinical presentation paramedics noted that he appeared “drunk” but had nor- Methanol is widely used in industry and is found in many mal vital signs and oxygen saturation. After transport to North American households. Because of its low freezing hospital, he complained of worsening abdominal pain. At point, methanol is a common component in gas line an- that time, his blood pressure was 110/70 mm Hg, heart tifreeze, glass cleaner and windshield wiper fluid. Methanol rate 72 beats/min, temperature 36°C, respiratory rate causes intoxication, often associated with nausea, vomiting (RR) 24 breaths/min and oxygen saturation 97% on room and abdominal pain. After a latent period of 24 (range 1 to air. Thirty minutes later, his Glasgow Coma Scale score 72) hours, weakness and respiratory difficulty may occur.1

had fallen to 9 (E2/M4/V3) (E = eye opening, M = motor The latent period may be longer if ethyl alcohol is a co-in- response, V = verbal response), and his RR had increased gestant or shorter if the volume of methanol is large. Visual to 30 breaths/min, but there were no focal neurologic disturbances, described as “walking in a snowstorm,” are signs, and physical examination was otherwise unremark- common. Ocular examination reveals pupillary dilatation, able. Fifteen minutes later he began convulsing, and a loss of pupillary reflexes, and hyperemia and edema of the rapid sequence intubation was performed with midazo- optic disk. Coma, seizures and death from cardiorespiratory lam, fentanyl and succinylcholine. The patient then de- arrest may occur. Laboratory evaluation may reveal a veloped pulseless electrical activity, followed by 2 widened osmolar gap and severe anion gap metabolic aci- episodes of hypotensive bradycardia. After resuscitation dosis. A pH of less than 7.0 is not uncommon.2 with epinephrine, he received dopamine and transcutaneous pacing while preparations were made for a transve- Pathophysiology nous pacemaker. The degree of toxicity correlates with the amount of Laboratory results were as follows: sodium (Na+) 153 methanol ingested, but not with presenting methanol lev- mmol/L, potassium (K+) 5.4 mmol/L, chloride (Cl–) 108 els. Latency between ingestion and toxicity occurs because – mmol/L, bicarbonate ions (HCO3 ) 5 mmol/L , blood urea of the time required to convert methanol to toxic metabo- nitrogen 5.9 mmol/L, creatinine 174 mmol/L and glucose lites. The toxic effects become apparent when ADH has 6 mmol/L. Arterial blood gas results were as follows: pH metabolized methanol to formaldehyde (Fig. 1). Formalde-

6.49, HCO3 5 mmol/L and partial pressure of carbon dioxide (PCO2) 62 mm Hg. Serum osmolarity was 487 mOsm. Methanol The anion and osmolar gaps were 40 and 169 mEq/L re- spectively. The severe anion gap metabolic acidosis and widened Alcohol dehydrogenase osmolar gap suggested toxic alcohol ingestion; therefore an ethanol infusion was started. Despite aggressive fluid resuscitation, 8 ampules of sodium bicarbonate, pressor support and transcutaneous pacing, the patient remained Formaldehyde hypotensive, which precluded dialysis. Asystolic arrest oc- curred, and care was withdrawn. Serum methanol levels were positive, and review of a head CT scan done before Aldehyde dehydrogenase arrest revealed bilateral globus pallidus ischemia.

Discussion Formic acid

Methanol, ethylene glycol and isopropyl alcohol are toxic alcohols that may be ingested accidentally or consumed as Folate ethanol substitutes. Like ethanol, all 3 cause intoxication and are metabolized by alcohol dehydrogenase (ADH), a process that creates toxic metabolites. Methanol and ethyl- CO2 + H2O ene glycol produce the most severe and life-threatening poisonings. Fig. 1. Methanol metabolism

January • janvier 2002; 4 (1) CJEM • JCMU 35 Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.93, on 26 Sep 2021 at 20:37:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1481803500006035 Henderson and Brubacher

hyde is highly toxic but is rapidly degraded by aldehyde after ingestion), patients exhibit signs of intoxication, stu- dehydrogenase and other nonspecific enzymes to formic por, nausea and vomiting, hallucinations and seizures. The acid,3 which is responsible for the metabolic acidosis and cardiorespiratory phase (12 to 24 hours after ingestion) is anion gap. Further metabolism of formic acid to carbon heralded by the onset of hypotension, tachypnea, conges- dioxide is dependent on folate.4 tive heart failure or, occasionally, myositis. The renal stage Visual changes with methanol poisoning are due to micro- (24 to 72 hours after ingestion) is marked by flank pain and tubule and mitochondrial destruction in the retrolaminar op- oxalate crystalluria, followed by the development of olig- tic nerve.5 Survivors may also develop a parkinsonism-like uric renal failure that may necessitate long-term dialysis.4 syndrome, which correlates with CT evidence of destruction in the putamen and subcortical white matter hemorrhage.6 Pathophysiology ADH converts ethylene glycol to glycoaldehyde (Fig. 2), Ethylene glycol which is rapidly converted to glycolic acid.7 Glycolic acid is slowly converted to glyoxylic acid, which is then de- Clinical presentation graded via several pathways. The major pathway is gly- Ethylene glycol is a component of radiator antifreeze, colic acid to oxalate, and the next most important involves coolants, polishes and cleansers. It is relatively innocuous the creation of glycine via pyridoxine-dependent amino- until metabolized to toxic breakdown products, which ex- transferases. Hypocalcemia and severe metabolic acidosis plains its 4- to 12-hour latent period.1 Ethylene glycol toxic- are common laboratory findings, and glycolic acid is the ity is divided into 3 distinct phases: central nervous system metabolite most responsible for the anion gap metabolic (CNS) depression, cardiorespiratory toxicity and renal toxi- acidosis. Oxalate crystal deposition in tissues is another city. It is a more potent CNS depressant than methanol and, mechanism of toxicity, and oxalate crystalluria is a hall- during the CNS depressant phase (30 minutes to 12 hours mark of ethylene glycol poisoning.8 Pyridoxine administra-

Ethylene glycol

Alcohol dehydrogenase

Glycoaldehyde

Aldehyde dehydrogenase

Glycolic acid Lactate dehydrogenase and glycolic acid oxidase

Glyoxylic acid

Pyridoxine Thiamine

Glycine Oxalate acid Formic acid α – hydroxy – β - ketoadipate

Fig. 2. Ethylene glycol metabolism

36 CJEM • JCMU January • janvier 2002; 4 (1) Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.93, on 26 Sep 2021 at 20:37:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1481803500006035 Methanol and ethylene glycol poisoning

tion may shift glycolic acid metabolism away from the source of anion gap metabolic acidosis can be quickly production of oxalate and toward the production of ruled out by measuring lactate, ketones and renal function. glycine, which is less toxic. One major exception is that increased lactate levels can also be due to an extrinsic toxin causing cellular hypoxia, Diagnosis of ethylene glycol and methanol poisoning as occurs in carbon monoxide poisoning. In the absence of Rapid recognition and treatment of toxic alcohol poison- an intrinsic cause, anion gap metabolic acidosis suggests ings is crucial to reduce the occurrence of morbidity and extrinsic poisoning. In this case, a toxin (e.g., acetylsali- mortality. The key to diagnosis is a high index of suspicion cylic acid) or a toxic metabolite (e.g., glycolic acid) may and a thorough history in patients who appear drunk or be the unmeasured anion (Table 2). who have unexplained acid–base abnormalities. Unfortu- nately, the early findings, including nausea, vomiting and = altered mentation, are nonspecific. Features such as hyper- Total cations Total anions pnea (compensation for metabolic acidosis), visual com- plaints, pupillary dilation and a latent period between inebriation and more severe symptoms are suggestive of Unmeasured cations methanol poisoning, whereas calcium oxalate crystalluria, present in 50% of ethylene glycol poisonings, is consid- Unmeasured Anion ered strong enough evidence to begin treatment in appro- anions priate circumstances. Urine that appears fluorescent under gap a Wood’s lamp may be diagnostic, since fluorescein dye is often added to ethylene glycol in radiator antifreeze to help detect radiator leaks. Fluorescein is not added to other compounds containing ethylene glycol, so the absence of Chloride fluorescence does not rule out ethylene glycol poisoning. Sodium

The anion gap Because serum is electroneutral, the sum of the positively Bicarbonate charged particles (cations) must equal the sum of the nega- tively charged particles (anions). The routinely measured cations are sodium and potassium and the routinely measured anions are chloride and bicarbonate. The difference Fig. 3. The anion gap is the difference between the mea- between the measured cations and the measured anions is sured cations and the measured anions. It is also equal to unmeasured anions minus unmeasured cations. For the pur- known as the anion gap. It represents unmeasured anions, pose of the calculation, potassium is considered an un- such as phosphates, sulfates, albumin and organic acids, measured cation. and unmeasured cations, such as calcium and magnesium. The anion gap actually refers to unmeasured anions – un- Table 1. Intravenous and oral ethanol doses10 measured cations (Fig. 3). Infusion rate, By convention, potassium is not included in anion gap mL/kg per hour calculations, so the anion gap is defined as follows: anion Loading Patient Patient not gap = sodium – (chloride + bicarbonate). When defined in Form of ethanol dose, undergoing undergoing this way the normal anion gap is 7 ± 4 mEq/L.9 An in- administration* mL/kg dialysis dialysis crease in the anion gap is usually due to unmeasured acids 10% ethanol IV and is known as anion gap metabolic acidosis. Unmea- Nondrinker 7.6 2.1 0.8 sured anions that cause anion gap acidosis may be intrinsic Chronic drinker 7.6 3.3 2.0 or extrinsic (Table 1). Intrinsic anions include lactate, ke- 43% ethanol by mouth† tones and the organic acids that accumulate in renal failure. Nondrinker 1.8 0.5 0.2 Lactic acidosis is caused by cellular hypoxia from hypop- Chronic drinker 1.8 0.8 0.5 erfusion, toxins that interfere with cellular metabolism or, *Ethanol doses assume a target serum ethanol level of 100 mg/dL and a dialysis rarely, inborn errors of metabolism. Ketones accumulate in time of 6 hours. diabetic, starvation or alcoholic ketoacidosis. An intrinsic †Ethanol is usually diluted to 20% ethanol by volume.

January • janvier 2002; 4 (1) CJEM • JCMU 37 Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.93, on 26 Sep 2021 at 20:37:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1481803500006035 Henderson and Brubacher

Every widened anion gap should be evaluated. If the out” toxic alcohol exposure.11,12 Unfortunately, there are cause is transient lactic acidosis, such as occurs after a problems with this approach, and an osmolar gap within seizure, it is reasonable to follow the laboratory abnormali- the normal range does not necessarily rule out significant ties to resolution. If the anion gap is unexplained, then sali- toxic alcohol poisoning. cylate, methanol and ethylene glycol levels should be measured. In many institutions it is often impossible to perform Caught in the osmolar gap these assays rapidly enough for the results to be useful in The normal range for the osmolar gap is wide. One group clinical decision-making. Thus, an elevated anion gap not of investigators found that the mean normal osmolar gap explained by lactate, ketones, renal failure or salicylates was –2 (standard deviation 6.1) mEq/L.13 Assuming that may be considered an indication for empiric treatment for each milliequivalent per litre of ethylene glycol contributes toxic alcohol poisoning, pending definitive diagnosis. 1 to the osmolar gap, a patient with an osmolar gap of –5 and a toxic ethylene glycol level of 10 mEq/L (64 mg/dL) Osmolarity and the osmolar gap would have an osmolar gap of 5 mEq/L which would be Osmolarity is defined as the number of particles in a litre considered “normal” if 10 mEq/L was considered the up- of solution. Serum osmolarity can be approximated by per limit of normal. Furthermore, because the metabolites adding the molar concentrations of the most common con- of ethylene glycol and methanol do not contribute to the stituents of serum. The most commonly used formula for osmolar gap, the gap may become normal late in these poi- calculating osmolarity is as follows: calculated osmolarity sonings, once the primary agent has been metabolized. = 2 Na + urea + glucose + ethanol (measured as milli- Another problem is that different formulae may be used equivalents per litre). to calculate osmolarity, and their results may differ sub- The osmolar gap (OG) is the difference between mea- stantially.13 Some laboratories still measure osmolarity by

sured osmolarity and calculated osmolarity: OG = Om – Oc. the vapour pressure method, despite its potential inaccu- This difference is normally accounted for by calcium, racy.14 Finally, a widened osmolar gap may occur in lactic lipids and proteins, but it may be increased by exogenous acidosis and ketoacidosis.15,16 For all these reasons, some compounds such as glycols and smaller alcohols,8 includ- authors feel that osmolar gap is of little use in differentiat- ing methanol, ethylene glycol, mannitol, glycerol and iso- ing the causes of anion gap metabolic acidosis.17 It is most propyl alcohol. Of these, only methanol and ethylene gly- important to re-emphasize that a normal osmolar gap does col commonly cause severe metabolic acidosis with an not rule out a toxic alcohol ingestion. elevation of both the anion and osmolar gaps.8 Clinicians often believe that an osmolar gap of less than 10 “rules Other tests Urine microscopy will reveal calcium oxalate crystals in Table 2. Causes of increased anion gap metabolic acidosis 50% of ethylene glycol poisoned patients at admission,8 Endogenous causes and this figure increases with time. Methanol poisoning Lactate: in shock states, following seizures, as a result of can be confirmed by determination of serum methanol cellular toxins level, a test that is available in most large hospital laborato- Ketones: in diabetic ketoacidosis, starvation ketoacidosis ries that use gas chromatography. Ethylene glycol determi- or alcoholic ketoacidosis nation is less widely available, but elevated urine and Organic acids: in renal failure serum oxalate levels are helpful adjuncts. Exogenous causes* Salicylates: weak acids responsible for acidosis and anion gap Management of ethylene glycol Nonsteroidal anti-inflammatory agents: weak acids that and methanol poisoning cause an anion gap The management of ethylene glycol and methanol poison- Methanol: anion gap acidosis caused by metabolites (specifically formic acid) ings is similar. Resuscitation, stabilization and decontami- Ethylene glycol: anion gap acidosis caused by metabolites nation are the initial goals. Gastric lavage, with a standard Toluene: anion gap acidosis caused by metabolites nasogastric tube rather than the potentially more dangerous (specifically hippuric acid) Ehrlich tube, may be of benefit within 1 to 2 hours of inges- *Many toxins, including cyanide, phenformin, iron and isoniazide, cause tion. Beyond this, rapid absorption renders the risk of aspi- anion gap acidosis through accumulation of endogenous anions such as lactate. The most common causes of anion gap metabolic acidosis are keto- ration greater than the possible benefits. Activated charcoal acidosis, lactic acidosis and salicylates. If these possibilities have been exclud- 18 ed, toxic alcohol poisoning (with methanol or ethylene glycol) should be is not helpful unless a poly-drug ingestion is suspected. considered. Dextrose, oxygen, naloxone and thiamine should be

38 CJEM • JCMU January • janvier 2002; 4 (1) Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.93, on 26 Sep 2021 at 20:37:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1481803500006035 Methanol and ethylene glycol poisoning

given to obtunded patients. Forced diuresis does not sub- peritoneal dialysis.25 Indications for hemodialysis in con- stantively change the rate of excretion and presents the risk firmed overdose include metabolic acidosis, renal compro- of precipitating pulmonary edema — particularly in pa- mise, visual symptoms (with methanol), or serum concen- tients who are developing renal compromise.19 The 4 major trations above 4.03 mmol/L for ethylene glycol and above goals in the treatment20 of ethylene glycol and methanol 7.8 mmol/L for methanol.26 poisonings are as follows: Alkalinization 1) inhibition of ADH to prevent toxic metabolite forma- Methanol and ethylene glycol metabolites may generate tion, hundreds of milliequivalents of excess acid per hour.27 Un- 2) correction of the acidosis with bicarbonate, like lactic acid, these acid metabolites are not degraded to 3) use of specific enzymatic cofactors such as folate, thi- bicarbonate, and the result is severe metabolic acidosis that amine and pyridoxine to modify deleterious metabolic is “bicarbonate resistant.” Sodium bicarbonate is indicated pathways, and if the serum pH falls below 7.2. Bicarbonate may help re- 4) removal of the toxin and metabolites by hemodialysis. move formic acid from the CNS and increase its renal excretion by “ion trapping.” Massive amounts may be neces- ADH blockade sary, although the risks of hypernatremia and pulmonary The affinity of ethanol for ADH is 100 times that of ethyl- edema must be kept in mind. ene glycol and 10 to 20 times that of methanol.1 As a result, ethanol, given orally or parenterally to achieve con- Co-factors centrations of 20 to 30 mmol/L (100 to 150 mg/dL), will In cases of ethylene glycol poisoning, thiamine and pyri- saturate ADH binding sites and prevent it from metaboliz- doxine may decrease oxalic acid formation and shift me- ing ethylene glycol or methanol.18 Once the ADH receptors tabolism to less toxic metabolites. Recommended doses are saturated, serum alcohol levels should be monitored to for both thiamine and pyridoxine are 100 mg administered ensure that therapeutic levels are maintained. Higher doses intravenously (IV), daily.28 In methanol intoxication, the are necessary in alcoholic patients and those undergoing degradation of formic acid to carbon dioxide (Fig. 1) is fo- hemodialysis (Table 1).10 late dependent. There is evidence that folate (50 mg IV Indications for ethanol therapy include a history, clinical q4h) may enhance the elimination of formic acid, decreas- picture or laboratory evidence suggestive of methanol or ing toxicity.29 ethylene glycol poisoning. It is often prudent to begin therapy before a definitive diagnosis is made. Prognostic factors after ingestion of toxins 4-Methylpyrazole (4-MP or fomepizole) is a newly re- Liu and colleagues30 found that coma or seizures at presen- leased antidote that reversibly inhibits ADH. 4-MP is tation and a serum pH less than 7.0 were the factors most rapidly effective, can be administered orally or parenter- closely correlated with death. These authors also found ally, does not cause the inebriation or hypoglycemia seen that prolonged acidosis was associated with increased neu- with ethanol, and has not been associated with serious ad- rologic sequelae and that shorter time to dialysis did not verse events.8 A prospective evaluation showed that par- improve survival; however, the latter finding may have enteral 4-MP, given as a 15 mg/kg loading dose followed been due to a selection bias whereby the sickest patients by 10 mg/kg every 12 hours for 48 hours, maintained ther- underwent dialysis most urgently.30 apeutic plasma 4-MP levels (≥10 µmol/L) and was safe and effective for ethylene glycol poisoning.21 Although Pregnancy prospective trials have not yet been published,22 a recent Limited anecdotal evidence is available to guide the ther- case series suggested that fomepizole is a reasonable treat- apy of pregnant patients.31 Of note, there is no evidence ment alternative in methanol poisonings.23 There is some that standard ethanol therapy adversely affects fetal out- evidence to suggest that 4-MP may decrease or eliminate come. the need for dialysis.24 Conclusions Dialysis Once ADH is blocked, the half-life of methanol and ethyl- Methanol and ethylene glycol poisonings are common and ene glycol become prolonged. Removal is accomplished serious. Ethanol treatment must often begin before the di- by means of hemodialysis, which is far more effective than agnosis is established. While often useful, both the anion

January • janvier 2002; 4 (1) CJEM • JCMU 39 Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.93, on 26 Sep 2021 at 20:37:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1481803500006035 Henderson and Brubacher

gap and the osmolar gap can be falsely reassuring in at- cation by osmometry. Arch Intern Med 1986;146:1843-4. tempts to rule out these toxins. Dialysis is the definitive 15. Schelling JR, Howard RL, Winter SD, Linas SL. Increased os- treatment in confirmed poisoning and should be used in molal gap in alcoholic ketosis and lactic acidosis. Ann Intern conjunction with ethanol therapy. Cofactor strategies (fo- Med 1990;113:580-2. late, thiamine and pyridoxine) may offer benefit. Recent 16. Davidson DF. Excess osmolal gap in diabetic ketoacidosis ex- evidence supports the use of 4-MP in methanol and ethyl- plained. Clin Chem 1992;38:755-7. ene glycol poisoning. 17. Glaser DS. Utility of the serum osmolal gap in the diagnosis of methanol or ethylene glycol ingestion. Ann Emerg Med Competing interests: None declared. 1996;27:343-6.

18. Ethanol for ethylene glycol poisoning [letter]. N Engl J Med References 1981;304:976-8.

1. Goldfrank LR, Flomenbaum NE. Goldfrank’s toxicologic emer- 19. Catchings TT, Beamer WC, Lundy L, Prough DS. Adult respira- gencies, 6th ed. Stamford (CT): Appleton & Lange; 1998. p. tory distress syndrome secondary to ethylene glycol ingestion. 1049-60. Ann Emerg Med 1985;14:594-6.

2. Jacobsen D, Hewlett TP, Webb R, Brown ST, Ordinaro AT, 20. Jacobsen D, McMartin KE. Methanol and ethylene glycol poi- McMartin KE. Ethylene glycol intoxication: evaluation of kinet- sonings: mechanism of toxicity, clinical course, diagnosis and ics and crystalluria. Am J Med 1988;84:145-52. treatment. Med Toxicol 1986;1:309.

3. McMartin KE, Martin-Amat G, Noker PE, Tephly TR. Lack of 21. Brent J, McMartin K, Phillips S, Burkhart KK, Donovan JW, a role for formaldehyde in methanol poisoning in the monkey. Wells M, et al. Fomepizole for the treatment of ethylene glycol Biochem Pharmacol 1979;28:645-9. poisoning. N Engl J Med 1999;340:832-8.

4. McMartin KE, Martin-Amat G, Makar AB, Tephly TR. 22. Burns MJ. Treatment of methanol poisoning with intravenous 4- Methanol poisoning. V. Role of formate metabolism in the mon- methylpyrazole. Ann Emerg Med 1997;30:829-32. key. J Pharmacol Exp Ther 1977;201:564-72. 23. Brent J, McMartin K, Phillips S, Aaron C, Kulig K. Fomepizole 5. Sharpe JA, Hostovsky M, Bilbao JM, Rewcastle NB. Methanol for the treatment of methanol poisoning. N Engl J Med 2001; optic neuropathy: a histopathological study. Neurology 1982;32: 344:424-9. 1093-100. 24. Jacobsen D, Sebastian CS, Barron SK, Carriere EW, McMartin 6. Aquilonius SM, Bergstrom K, Enoksson P, Hedstrand U, Lund- KE. Effects of 4-methylpyrazole, methanol/ethylene glycol anti- berg PO, Mostrom U, et al. Cerebral computed tomography in dote, in healthy humans. J Emerg Med 1990;8:455-61. methanol intoxication. J Comput Assist Tomogr 1980;4:425-8. 25. Keyvan-Larijarni H, Tannenberg AM. Methanol intoxication. 7. Hewlett TP McMartin KE, Lauro AJ, Ragan FA. Ethylene gly- Comparison of peritoneal dialysis and hemodialysis treatment. col poisoning. The value of glycolic acid determinations for di- Arch Intern Med 1974;134:293-6. agnosis and treatment. J Toxicol Clin Toxicol 1986;24:389-402. 26. Gonda A, Gault H, Churchill D, Hollomby D. Hemodialysis for 8. Jacobsen D, McMartin KE. Antidotes for methanol and ethylene methanol intoxication. Am J Med 1978:64:749-58. glycol poisoning. Clin Toxicol 1997;35:127-43. 27. Michelis MF, Mitchell B, Davis BB. “Bicarbonate resistant” 9. Winter SD, Pearson R, Gabow PA, Schultz AL, Lepoff RB. The metabolic acidosis in association with ethylene glycol intoxica- fall of the serum anion gap. Arch Intern Med 1990;150:311-3. tion. Clin Toxicol 1976;9:53-60.

10. McCoy HG, Cipolle RJ, Ehlers SM, Sawchuk RJ, Zaske DE. 28. Birnbaumer DM, Bessen HA. Other alcohols. In: Rosen P, Severe methanol poisoning: an application of a pharmacokinetic Barkin R, editors. Emergency medicine concepts and clinical model for ethanol therapy and hemodialysis. Am J Med practice, 4th ed. St Louis: CV Mosby; 1998. p. 2520-33. 1979;67:804-7. 29. From the NIH: Use of folate analogue in treatment of methyl al- 11. Smithline N, Gardner KD Jr. Gaps: anionic and osmolal. JAMA cohol toxic reactions is studied. JAMA 1979;242:1961-2. 1976;236:1594-7. 30. Liu JL, Daya MR, Carrasquillo O, Kales SN. Prognostic factors in 12. Genari FJ. Serum osmolality: uses and limitations. N Engl J patients with methanol poisoning. Clin Toxicol 1998;36:175-81. Med 1984;310:102-5. 31. Hantson P, Lambermont JY, Mahieu P. Methanol poisoning 13. Hoffman RS, Smilkstein MJ, Howland MA, Goldfrank LR. Os- during late pregnancy. Clin Toxicol 1997;35:187-91. mol gaps revisited: normal values and limitations. J Toxicol Clin Toxicol 1993;31:81-93. Correspondence to: Dr. William Henderson, Department of Emergency 14. Walker JA, Schwartzbard A, Krauss EA, Sherman RA, Eisinger Medicine, Vancouver Hospital and Health Sciences Centre, 855 W 12th RP. The missing gap. A pitfall in the diagnosis of alcohol intoxi- Ave., Vancouver BC V5Z 1M9; fax 604 737-1959; [email protected]

40 CJEM • JCMU January • janvier 2002; 4 (1) Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.93, on 26 Sep 2021 at 20:37:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1481803500006035